> The outcome vectors aren't normalized right? I'm not sure about it, maybe they are, it wouldn't be strange > So there could be a hefty amount of difference

> The outcome vectors aren't normalized right? I'm not sure about it, maybe they are, it wouldn't be strange > So there could be a hefty amount of difference in this space? Maybe not on concept, but perhaps on image quality? Sure, each text could have more than one image matching the same representation (cosine wise), but maybe the changes wouldn't look much as "concepts" in the image but other features (sharpness, light, noise, actual pixel values, etc) It would be curious to check, definitely